The need for emotion monitoring for people undergoing anxiety disorders made me build the sentiment analysis part of the app. To make the app a complete mental health suite, I added a Feed that lists out articles pertaining to mental health and also added options to connect with Therapists through the application to allow for easier access to therapists.
What it does
Wellness has two parts -
1) Wellness Diary - The application for users in search of help
2) Wellness Therapist - The application for therapists who want to help
Wellness Diary: Wellness Diary provides a platform for users to monitor their emotions on a daily basis and also maintain a diary within the application. This could potentially enable users to track their responses to triggers and/or to therapy. It can, however, also be used just as a diary. Wellness Diary also provides a Feed which lists out articles about mental health. There are also options to search for therapists, connect with them and then send them messages and receive messages in an in-app Inbox. A user can send a therapist message only when he/she connects with the therapist.
Wellness Therapist: Wellness Therapist allows the therapist to set up and maintain a profile and interact with their patients who have chosen them through in-app messages.
How I built it
I built both the android applications using Java. A Parse backend deployed on Heroku and local SQLite databases support both the applications.
Wellness Diary makes use of Microsoft's Cognitive Service APIs to perform sentiment analysis. The therapists and messages are obtained from the Parse backend. Articles are also fed to the Parse backend and can be obtained by web scraping.
Wellness Therapist uses the Parse backend to retrieve messages and also enable therapists to send messages to their patients.
Challenges I ran into
Setting up the backend while allowing for foreign key constraints and unique values in Parse. In an SQL based backend system, setting up such constraints is easy but with Parse, such constraints have to be coded into the server code in Node.js.
What I learned
I learned how to write code so that there is efficient communication between the mobile application and the Parse backend.
What's next for Wellness Diary
Monetization - Users would connect to therapists only after paying a fee based on the experience and a "rating" that a therapist holds. In-App communication - Wellness will contain a more robust communication system within the application that allows for easier messaging as well as video and phone calls.
All the code for this application is in a private repository on GitHub. I will be willing to provide access to the judges if required. Please email me at firstname.lastname@example.org for access to the repository.